Strong Signals --> Ditch Shopify CAPI

Thought of doing it with my favorite analogies to get the crux. 

Running ads with bad signals is like ordering pizza without giving your address—great effort, but it’s never getting to the right place. Fix your signal and deliver those sales hot and fresh. 

So hello-hello again. 

Let’s talk about signals today. 

Basically what are signals?  

Signals are actions or behaviors users take online and offline that help Meta spot potential buyers. Without these signals, Meta would be like trying to find a needle in a haystack—blindfolded.

“Looks like brands are almost losing 200,000 dollars on ads by giving bad signals”

What we see most of the brands use Shopify CAPI and the quality (EMQ) of the signals looks something like this

In the world of modern marketing, these are just not expected as the fundamentals. 

Cracking the basics: 

High event quality is crucial as it helps Meta accurately identify users, ensuring ads are delivered to the right audience.

Conversion API (CAPI) can improve event quality by utilizing server-side data, such as data from Shopify, to better attribute conversions and lower costs per action.

Basics? Done. But Meta Demands More! Go Beyond Old-School Signal Optimization.

What if I say that, we could optimize the campaign for the specific business goals of ours? Let’s say?

I need more money, so get me all high AOVs,
Launched new categories, need more of these.
No low AOVs, keep mid AOV in my eye,
Off-season slump, but maintain my ROI.

You can actually train the Meta algorithm for your business goal and get them the most. 

You know what just Meta announced? 

Meta is giving the media buyers the ability to connect their other analytics tools i.e. third-party tools to connect with the ad platforms. 

“We’re making changes to improve the system so we can get to a place where we’re providing customized optimizations for every individual advertiser”

Yeah people, Meta is figuring out custom conversion optimizations based on the business goals and needs. 

Just relying on the Facebook’s data is definitely not helping as Meta itself receives the leftovers 😛

Each campaign is an AI model itself

Meta ad algorithm is a machine learning algorithm. It needs to be trained on what kind of purchase does businesses want. Based on the user behavior learning, Meta can predict the ROAS that each user can provide which will help in maximizing the value and number of campaign conversions.

If your business goal is to get more high AOVs or mid AOVs? 

You already have the First-Party Data OPs foundation laid as the red carpet. Now, you have the liberty and freedom to work on them the fullest. 

If your business goal is to get more category based purchases? 

Let’s prove that it is just not theory? It is pure logic. 

This table is a clear proof of concept that when you align your campaign optimization with specific business goals, Meta’s algorithm delivers exactly what you’re aiming for.

For instance, in Campaign 1, the optimization goal was mid_AOV_men. As a result, this campaign achieved the highest percentage of mid_AOV_men purchases compared to others.

Similarly, in Campaign 2, the goal was low_AOV_men, and it delivered exactly that, outperforming other campaigns in attracting low AOV men purchases.

This is the true power of proper algorithm training—it manifests the desired results by precisely targeting the audience you need.

I think this is overboard for today. 

On the next newsletter I’m gonna talk about another Meta’s announcement and how 1PD OPs 3% club is already ready for it. 

Need a sneak peak?

Okay. Byeeeeeee … 

Welcome to 3% Marketers Club - Join here.